© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 4-GEMMASCOPE-TRANSCODER-16K
    7. 2540
    Prev
    Next
    INDEX
    Explanations

    a mix of truncated code snippets and instances of the word "arrested"

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ClientSize
    -0.71
     defaultstate
    -0.65
    StructEnd
    -0.64
    ContentAsync
    -0.61
    ModelSerializer
    -0.58
    GEBURTSDATUM
    -0.57
     beginnetje
    -0.56
    InjectAttribute
    -0.56
     समीक्षाओं
    -0.56
    featureID
    -0.56
    POSITIVE LOGITS
    Picking
    0.88
     CATCH
    0.86
     pick
    0.84
     catch
    0.83
     picked
    0.82
    pick
    0.82
     catches
    0.80
     Picking
    0.80
    Pick
    0.79
     picking
    0.79
    Activations Density 0.962%

    No Known Activations